Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A power-subsystem-monitoring-based computing system, comprising: a power subsystem; a first computing component that is coupled to the power subsystem; and a throttling controller that is coupled to the power subsystem and the first computing component, wherein the throttling controller is configured to: monitor a power draw via the power subsystem; activate a first throttling of the first computing component when the power draw via the power subsystem exceeds a power subsystem maximum power consumption that is determined using an identifier for the power subsystem, and deactivate the first throttling of the first computing component when the power draw via the power subsystem no longer exceeds the power subsystem maximum power consumption; activate a second throttling of the first computing component when the power draw via the power subsystem exceeds a power subsystem rated power consumption for a first time period, and deactivate the second throttling of the first computing component when the power draw via the power subsystem no longer exceeds the power subsystem rated power consumption; and reduce the operating capabilities of the first computing component when the second throttling has been performed for more than a second time period, and increase the operating capabilities of the first computing component when the second throttling has been performed for less than the second time period and the first computing component is operating below a first computing component predetermined operating capability.
A computing system monitors power consumption to manage performance and prevent overheating or damage. The system includes a power subsystem, a computing component, and a throttling controller. The controller monitors power draw from the power subsystem and applies dynamic throttling based on power thresholds. If power consumption exceeds a maximum limit (determined by the power subsystem's identifier), the controller activates temporary throttling to reduce the computing component's performance until power draw falls below the limit. Additionally, if power consumption exceeds a rated limit for a sustained period, the controller applies a second throttling mechanism. If this second throttling persists beyond a predefined duration, the system permanently reduces the computing component's operating capabilities. Conversely, if throttling is infrequent and the component operates below its maximum capacity, the system gradually restores its performance. This approach ensures system stability by balancing power efficiency and performance while preventing long-term degradation from excessive power usage.
2. The system of claim 1 , wherein the first computing component is a graphics processing system.
This technical summary describes a system for processing graphical data using a specialized computing component. The system addresses the challenge of efficiently handling complex graphical computations, which are increasingly demanded by modern applications such as gaming, virtual reality, and real-time simulations. Traditional central processing units (CPUs) often struggle with the parallel processing requirements of graphical workloads, leading to bottlenecks and inefficiencies. The system includes a first computing component, which is a graphics processing system (GPS), designed to accelerate graphical computations. The GPS is optimized for parallel processing, enabling it to handle multiple graphical operations simultaneously. This improves performance and reduces latency in applications requiring real-time rendering. The system also includes a second computing component, which may be a general-purpose processor or another specialized unit, to manage non-graphical tasks or coordinate with the GPS. The GPS may include dedicated hardware such as shaders, texture mapping units, and rasterization engines to enhance graphical processing efficiency. It may also support advanced features like ray tracing, tessellation, and programmable pipelines to deliver high-quality visual outputs. The system ensures seamless integration between the GPS and other components, allowing for efficient data transfer and synchronization. By leveraging a dedicated graphics processing system, the invention improves computational efficiency, reduces power consumption, and enhances the overall performance of graphical applications. This solution is particularly beneficial in environments where real-time rendering and high-resolution graphics are critical.
3. The system of claim 1 , further comprising: a second computing component that is coupled to the power subsystem and the throttling controller, wherein the throttling controller is configured to: activate a third throttling of the second computing component when the power draw via the power subsystem exceeds the power subsystem rated power consumption for a second time period immediately following the first time period, and deactivate the third throttling of the second computing component when the power draw via the power subsystem no longer exceeds the power subsystem rated power consumption.
The invention relates to power management in computing systems, specifically addressing the problem of managing power consumption to prevent system overload or damage when power draw exceeds rated limits. The system includes a power subsystem that monitors power consumption and a throttling controller that regulates the performance of computing components to maintain safe operating conditions. The system further includes a second computing component connected to the power subsystem and the throttling controller. The throttling controller is configured to activate a third throttling mechanism for the second computing component when power draw exceeds the rated power consumption for a second consecutive time period following an initial period of excess power draw. This throttling reduces the performance of the second computing component to lower power consumption. Once the power draw returns to within safe limits, the throttling controller deactivates the third throttling, allowing the second computing component to resume normal operation. This approach ensures that power consumption remains within safe thresholds while minimizing unnecessary performance degradation. The system dynamically adjusts throttling based on real-time power monitoring to balance performance and power efficiency.
4. The system of claim 3 , wherein the throttling controller is configured to: reduce the operating capabilities of the second computing component when the second throttling has been performed for more than a second time period, wherein the operating capabilities of the first computing component are reduced when the second throttling has been performed for more than the second time period and the operating capabilities of the second computing component have been reduced to a minimum level; and increase the operating capabilities of the second computing component when the second throttling has been performed for less than the second time period, the first computing component is operating at the first computing component predetermined operating capability, and the second computing component is operating below a second computing component predetermined operating capability.
A system for managing thermal and power constraints in computing devices involves dynamically adjusting the operating capabilities of multiple computing components to prevent overheating or excessive power consumption. The system includes a throttling controller that monitors the performance of at least two computing components, such as a central processing unit (CPU) and a graphics processing unit (GPU). When one of the components exceeds thermal or power limits, the system applies throttling to reduce its performance. If this throttling persists for more than a predefined second time period, the system further reduces the operating capabilities of the second computing component. Additionally, if the second component's performance is already at a minimum level, the system also reduces the performance of the first computing component. Conversely, if the throttling duration is less than the second time period, the system increases the operating capabilities of the second computing component, provided the first component is operating at its predetermined level and the second component is operating below its predetermined level. This adaptive approach ensures balanced performance while maintaining thermal and power efficiency.
5. The system of claim 3 , wherein the second computing component is a central processing system.
This invention relates to a distributed computing system designed to improve processing efficiency and resource allocation. The system addresses the challenge of optimizing computational tasks across multiple computing components, particularly when handling complex or resource-intensive operations. A central processing system (CPS) serves as a primary computing component, coordinating and managing tasks distributed among secondary computing components. The CPS is responsible for task allocation, load balancing, and ensuring efficient utilization of computational resources. Secondary computing components, which may include specialized processors or peripheral devices, execute tasks assigned by the CPS. The system dynamically adjusts task distribution based on real-time performance metrics, such as processing speed and resource availability, to enhance overall system efficiency. This architecture allows for scalable and flexible computing solutions, adaptable to varying workload demands. The invention aims to reduce processing bottlenecks and improve system responsiveness by leveraging the strengths of both central and distributed computing elements.
6. The system of claim 1 , wherein the throttling controller is configured to: characterize the power subsystem and determine the power subsystem maximum power consumption and the power subsystem rated power consumption.
A system for managing power consumption in electronic devices addresses the challenge of optimizing energy usage while maintaining performance. The system includes a power subsystem that supplies energy to various components and a throttling controller that regulates power delivery. The throttling controller is designed to characterize the power subsystem, assessing its operational limits. Specifically, it determines the maximum power consumption the subsystem can handle under peak conditions and the rated power consumption, which represents the sustained power level the subsystem is designed to operate at without degradation. By distinguishing between these two thresholds, the controller can dynamically adjust power delivery to prevent overheating, component damage, or system instability while ensuring efficient operation. This characterization allows the system to balance performance demands with power constraints, particularly in devices where thermal or power limitations could otherwise restrict functionality. The approach is useful in portable electronics, data centers, or any application where power management is critical to reliability and longevity.
7. The system of claim 1 , wherein the power subsystem is a power adapter.
The invention relates to a system for managing power distribution in electronic devices, particularly addressing the challenge of efficiently supplying and regulating power to multiple components within a compact device. The system includes a power subsystem designed to convert and distribute electrical power from an external source to various internal components, ensuring stable operation and energy efficiency. The power subsystem is specifically implemented as a power adapter, which converts alternating current (AC) from a wall outlet into direct current (DC) suitable for powering the device. This adapter may include features such as voltage regulation, current limiting, and protection against power surges or short circuits. The system also incorporates a control module that monitors power usage, adjusts distribution based on demand, and optimizes energy consumption. Additionally, the system may include a thermal management module to prevent overheating by dynamically adjusting power delivery or activating cooling mechanisms. The power adapter may be detachable or integrated, depending on the device's design requirements. The overall system ensures reliable power delivery while maintaining safety and efficiency in electronic devices.
8. An Information Handling System (IHS), comprising: a processing system; and a memory system that is coupled to the processing system and that includes instructions that, when executed by the processing system, cause the processing system to provide a throttling engine that is configured to: monitor a power draw via a power subsystem coupled to the processing system; activate a first throttling of a Graphics Processing Unit (GPU) that is coupled to the processing system when the power draw via the power subsystem that powers the GPU exceeds a power subsystem maximum power consumption that is obtained by performing a database lookup, and deactivate the first throttling of the GPU when the power draw via the power subsystem no longer exceeds the power subsystem maximum power consumption; activate a second throttling of the GPU when the power draw via the power subsystem exceeds a power subsystem rated power consumption for a first time period, and deactivate the second throttling of the GPU when the power draw via the power subsystem no longer exceeds the power subsystem rated power consumption; and reduce the operating capabilities of the GPU when the second throttling has been performed for more than a second time period, and increase the operating capabilities of the GPU when the second throttling has been performed for less than the second time period and the GPU is operating below a GPU predetermined operating capability.
This invention relates to power management in information handling systems (IHS), specifically for controlling power consumption of a graphics processing unit (GPU) to prevent overheating or exceeding power limits. The system includes a processing system and a memory system with instructions to implement a throttling engine. The engine monitors power draw via a power subsystem connected to the GPU. If the power draw exceeds a maximum power consumption threshold (retrieved via a database lookup), the GPU is throttled until the power draw falls below this threshold. Additionally, if the power draw exceeds a rated power consumption for a first time period, a second throttling mechanism is activated, which remains active until the power draw falls below the rated power level. If this second throttling persists for more than a second time period, the GPU's operating capabilities are reduced. Conversely, if the second throttling lasts less than the second time period and the GPU is operating below a predetermined capability, its operating capabilities are increased. This approach ensures the GPU operates within safe power limits while optimizing performance.
9. The IHS of claim 8 , wherein the throttling engine is configured to: activate a third throttling of a Central Processing Unit (CPU) that is coupled to the processing system when the power draw via the power subsystem exceeds the power subsystem rated power consumption for a second time period immediately following the first time period, and deactivate the third throttling of the CPU when the power draw via the power subsystem no longer exceeds the power subsystem rated power consumption.
This invention relates to power management in computing systems, specifically addressing the challenge of preventing excessive power draw that could damage hardware components. The system includes a processing system with a power subsystem that monitors power consumption and a throttling engine that regulates power usage. The throttling engine is designed to activate a third throttling mechanism for a Central Processing Unit (CPU) when the power draw exceeds the rated power consumption of the power subsystem for a second consecutive time period. This third throttling is deactivated once the power draw returns to acceptable levels. The system also includes a first throttling mechanism that activates when power draw exceeds the rated power consumption for a first time period and a second throttling mechanism that activates when power draw exceeds the rated power consumption for a second time period immediately following the first. The second throttling is deactivated when power draw no longer exceeds the rated power consumption. The third throttling, triggered by the second consecutive period of excessive power draw, provides an additional layer of protection to prevent hardware damage. The system ensures that power consumption remains within safe limits by progressively applying throttling measures based on the duration and frequency of power draw exceeding rated thresholds.
10. The IHS of claim 9 , wherein the throttling engine is configured to: reduce the operating capabilities of the CPU when the second throttling has been performed for more than a second time period, wherein the operating capabilities of the GPU are reduced when the second throttling has been performed for more than the second time period and the operating capabilities of the CPU have been reduced to a minimum level; and increase the operating capabilities of the CPU when the second throttling has been performed for less than the second time period, the GPU is operating at the GPU predetermined operating capability, and the CPU is operating below a CPU predetermined operating capability.
This invention relates to thermal management in computing systems, specifically for dynamically adjusting the performance of a central processing unit (CPU) and graphics processing unit (GPU) to prevent overheating. The system monitors thermal conditions and applies throttling to reduce performance when temperature thresholds are exceeded. A throttling engine controls the CPU and GPU based on the duration and frequency of throttling events. If throttling occurs more than a specified time period, the CPU's operating capabilities are reduced. If the CPU is already at its minimum performance level, the GPU's capabilities are also reduced. Conversely, if throttling occurs less frequently than the specified time period, the CPU's performance is increased, provided the GPU is operating at its predetermined level and the CPU is below its predetermined performance level. This approach balances thermal safety with performance optimization, ensuring sustained operation without excessive heat buildup. The system dynamically adjusts power states to maintain thermal thresholds while maximizing computational efficiency.
11. The IHS of claim 10 , wherein the throttling engine is configured to: characterize the power subsystem and determine the power subsystem maximum power consumption and the power subsystem rated power consumption.
A system and method for managing power consumption in an information handling system (IHS) addresses the challenge of optimizing power usage while maintaining performance. The invention focuses on dynamically adjusting power delivery to prevent overheating, system instability, or excessive energy consumption. A key component is a throttling engine that monitors and controls power distribution to various subsystems within the IHS. The throttling engine is configured to analyze the power subsystem, which includes components such as processors, memory, and storage devices. It determines the maximum power consumption the subsystem can handle without failure and the rated power consumption, which is the typical operating power under normal conditions. By comparing these values, the engine can identify potential power-related risks and adjust power delivery accordingly. This may involve reducing clock speeds, limiting power draw, or redistributing power to critical components. The system ensures that the IHS operates within safe power limits while maintaining efficiency. This is particularly useful in high-performance computing environments where power demands fluctuate frequently. The invention helps prevent thermal throttling, extends hardware lifespan, and improves energy efficiency without requiring manual intervention. The dynamic nature of the throttling engine allows for real-time adjustments, adapting to changing workloads and environmental conditions.
12. The IHS of claim 8 , wherein the power subsystem is a power adapter.
A power adapter for an information handling system (IHS) is designed to convert and regulate electrical power from an external source to supply the IHS. The power adapter includes a power conversion circuit that transforms input power, such as from an AC outlet, into a stable DC output suitable for the IHS. It may also incorporate power management features, such as voltage regulation, current limiting, and protection mechanisms to prevent damage from overvoltage, overcurrent, or short circuits. The adapter may be designed to interface with the IHS through a dedicated power connector, ensuring efficient power delivery while maintaining compatibility with the system's power requirements. This design allows the IHS to operate reliably across different power conditions, ensuring consistent performance and safety. The power adapter may also support additional features like power factor correction to improve energy efficiency and reduce harmonic distortion. By integrating these functions, the adapter provides a robust and efficient power solution for the IHS, ensuring stable operation in various environments.
13. The IHS of claim 8 , wherein the power subsystem is a battery.
This invention relates to an information handling system (IHS) designed to optimize power management, particularly in portable or battery-powered devices. The system addresses the challenge of maintaining reliable operation while efficiently managing power consumption, which is critical for extending battery life and ensuring uninterrupted functionality in mobile or remote environments. The IHS includes a power subsystem that is specifically configured as a battery, providing a portable and self-contained power source. This battery-powered subsystem enables the IHS to operate independently of external power sources, making it suitable for applications where continuous power availability is uncertain or impractical. The battery may be rechargeable, allowing for repeated use without the need for disposable power sources. The power subsystem is integrated with the IHS to supply electrical power to its components, including processing units, memory, and peripheral devices. The system may include power management features such as voltage regulation, current monitoring, and thermal management to ensure stable and efficient operation. These features help prevent over-discharge, overheating, or other conditions that could degrade battery performance or damage the system. Additionally, the IHS may incorporate power-saving modes, such as sleep or hibernation states, to further conserve battery life when the system is idle. The battery may also support fast-charging capabilities to minimize downtime during recharging. This invention is particularly useful in portable computing devices, IoT (Internet of Things) devices, and other systems where battery life and power efficiency are critical factors. By integrating a battery as the primary power source, the IH
14. A method for providing power-subsystem-monitoring-based computing, comprising: monitoring, by a throttling subsystem, a power draw of a power subsystem; activating, by the throttling subsystem, a first throttling of a first computing component when the power draw via the power subsystem that powers the first computing component exceeds a power subsystem maximum power consumption that is determined using an identifier for the power subsystem, and deactivating the first throttling of the first computing component when the power draw via the power subsystem no longer exceeds the power subsystem maximum power consumption; activating, by the throttling subsystem, a second throttling of the first computing component when the power draw via the power subsystem exceeds a power subsystem rated power consumption for a first time period, and deactivating the second throttling of the first computing component when the power draw via the power subsystem no longer exceeds the power subsystem rated power consumption; and reducing, by the throttling subsystem, the operating capabilities of the first computing component when the second throttling has been performed for more than a second time period, and increasing the operating capabilities of the first computing component when the second throttling has been performed for less than the second time period and the first computing component is operating below a first computing component predetermined operating capability.
This invention relates to power management in computing systems, specifically a method for dynamically adjusting computing component performance based on power subsystem monitoring. The system addresses the problem of ensuring stable operation while preventing excessive power draw that could damage power subsystems or trigger safety mechanisms. A throttling subsystem continuously monitors the power draw of a power subsystem supplying a computing component. If the power draw exceeds a predefined maximum power consumption threshold—determined using an identifier for the power subsystem—the system activates a first throttling mechanism to reduce the component's power consumption. This throttling is deactivated once the power draw falls below the threshold. Additionally, if the power draw exceeds a rated power consumption level for a sustained first time period, a second throttling mechanism is activated, which is deactivated once the power draw drops below the rated level. If the second throttling persists for a second time period, the system further reduces the computing component's operating capabilities to prevent long-term power overdraw. Conversely, if the second throttling duration is shorter than the second time period and the component is operating below its predetermined capability, the system increases its operating capabilities to optimize performance. This method ensures power subsystem safety while dynamically balancing performance and power efficiency.
15. The method of claim 14 , wherein the first computing component is a graphics processing system.
A graphics processing system is used to accelerate the execution of a software application by offloading computational tasks from a central processing unit (CPU). The system includes a first computing component, such as a graphics processing unit (GPU), that processes data in parallel to improve performance. The GPU receives input data from the CPU, executes parallel processing tasks, and returns processed data to the CPU. The system may also include a second computing component, such as a CPU or another GPU, that interacts with the first computing component to manage task distribution and synchronization. The GPU is optimized for handling large datasets and complex mathematical operations, making it suitable for applications like machine learning, scientific simulations, and real-time rendering. The system ensures efficient data transfer between the CPU and GPU, minimizing latency and maximizing throughput. This approach enhances overall system performance by leveraging the parallel processing capabilities of the GPU while maintaining compatibility with existing software architectures. The method ensures that tasks are distributed efficiently, reducing bottlenecks and improving computational efficiency.
16. The method of claim 14 , further comprising: activating, by the throttling subsystem, a third throttling of a second computing component when the power draw via the power subsystem that powers the second computing component exceeds the power subsystem rated power consumption for a second time period immediately following the first time period, and deactivating the third throttling of the second computing component when the power draw via the power subsystem no longer exceeds the power subsystem rated power consumption.
This invention relates to power management in computing systems, specifically addressing the problem of preventing power subsystem overload by dynamically throttling computing components when power consumption exceeds rated limits. The system includes a power subsystem that supplies power to multiple computing components and a throttling subsystem that monitors power draw. When the power draw exceeds the rated power consumption for a predefined first time period, the throttling subsystem activates a first throttling of a first computing component to reduce power consumption. Once the power draw returns to acceptable levels, the throttling is deactivated. If the power draw exceeds the rated power consumption again during a second time period immediately following the first, the throttling subsystem activates a second throttling of a second computing component, which is distinct from the first. This second throttling is deactivated once the power draw no longer exceeds the rated power consumption. The system ensures sustained operation by progressively throttling different components in response to repeated power overdraw events, preventing system instability or shutdown. The invention is particularly useful in high-performance computing environments where power management is critical.
17. The method of claim 16 , further comprising: reducing, by the throttling subsystem, the operating capabilities of the second computing component when the second throttling has been performed for more than a second time period, wherein the operating capabilities of the first computing component are reduced when the second throttling has been performed for more than the second time period and the operating capabilities of the second computing component have been reduced to a minimum level; and increasing, by the throttling subsystem, the operating capabilities of the second computing component when the second throttling has been performed for less than the second time period, the first computing component is operating at the first computing component predetermined operating capability, and the second computing component is operating below a second computing component predetermined operating capability.
This invention relates to a system for managing thermal and power constraints in computing devices by dynamically adjusting the operating capabilities of multiple computing components. The problem addressed is the need to prevent overheating or excessive power consumption while maintaining optimal performance, particularly in devices with multiple processing units or components. The system includes a throttling subsystem that monitors thermal and power conditions and applies throttling to reduce the operating capabilities of one or more computing components when thresholds are exceeded. The throttling subsystem further adjusts the operating capabilities of a second computing component based on the duration of throttling applied to it. If throttling of the second component exceeds a specified time period, the system reduces its operating capabilities and may also reduce the operating capabilities of a first computing component if the second component has already been throttled to its minimum level. Conversely, if throttling of the second component is below the specified time period, the system increases its operating capabilities, provided the first component is operating at its predetermined level and the second component is operating below its predetermined level. This ensures balanced performance and efficient resource utilization while preventing thermal or power-related failures.
18. The method of claim 16 , wherein the second computing component is a central processing system.
This invention relates to distributed computing systems, specifically methods for optimizing task allocation between multiple computing components to improve efficiency and performance. The problem addressed is the inefficient distribution of computational workloads in systems with diverse hardware components, leading to suboptimal resource utilization and performance bottlenecks. The method involves dynamically assigning tasks to different computing components based on their capabilities and current workload. A first computing component, such as a specialized accelerator (e.g., a graphics processing unit or AI accelerator), handles specific tasks optimized for its architecture. A second computing component, which is a central processing system (CPU), manages general-purpose tasks or those not suited for the first component. The system monitors performance metrics, such as processing speed and resource utilization, to determine the most efficient task distribution. If the first component is overloaded or a task is better suited for the CPU, the method reallocates the task to the CPU to balance the workload and prevent bottlenecks. This approach ensures that tasks are processed by the most appropriate hardware, improving overall system efficiency and performance. The dynamic allocation adapts to real-time conditions, allowing the system to handle varying workloads effectively.
19. The method of claim 14 , further comprising: characterizing, by the throttling subsystem, the power subsystem and determining the power subsystem maximum power consumption and the power subsystem rated power consumption.
A method for managing power consumption in electronic systems involves characterizing a power subsystem to determine its maximum and rated power consumption levels. The power subsystem includes components such as power supplies, voltage regulators, and power distribution networks. The method assesses the power subsystem's operational limits by analyzing its design specifications, thermal constraints, and load conditions. This characterization helps identify the highest power the subsystem can sustain without failure and the power level it is rated to handle under normal operating conditions. By distinguishing between these two values, the system can implement dynamic power management strategies, such as throttling or load balancing, to prevent overheating or component degradation. The method ensures efficient power utilization while maintaining system reliability. This approach is particularly useful in high-performance computing, data centers, and embedded systems where power efficiency and thermal management are critical. The characterization process may involve real-time monitoring, historical data analysis, or predictive modeling to accurately assess power consumption behavior under varying workloads.
20. The method of claim 14 , wherein the power subsystem is a power adapter.
A power management system for electronic devices addresses the challenge of efficiently distributing and regulating power from a power source to multiple components within a device. The system includes a power subsystem that converts and distributes power to various components, such as processors, memory, and peripheral devices, while ensuring stable operation and energy efficiency. The power subsystem may be integrated into the device or provided as an external power adapter. When implemented as an external power adapter, the subsystem converts incoming power from an external source, such as a wall outlet, into a form suitable for the device's internal components. It may include voltage regulation, current control, and protection mechanisms to prevent damage from power fluctuations or surges. The system dynamically adjusts power distribution based on the operational demands of the device, optimizing performance and battery life. This approach ensures reliable power delivery while minimizing energy waste, particularly in portable or high-performance devices where power efficiency is critical. The power adapter variant simplifies integration by handling external power conversion, reducing the complexity of internal power management circuits.
Unknown
September 1, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.